Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning — mentioned by mmmore

Wed, 01 Jan 2025 17:09:06 GMT

My understanding is that it's in vogue to use deep learning for complex control problems, and the results are fairly impressive. The idea is to train robotic motion end to end with RL. Not an expert so I don't know the strength and weaknesses versus classical approaches. https://blogs.nvidia.com/blog/eureka-robotics-research/ https://arxiv.org/abs/2108.10470

Paper and discussion history. A mention is not an endorsement.

Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning — mentioned by stonet2000

Tue, 24 Dec 2024 04:11:03 GMT

blogpost author here. This is not exactly the claim. I’m kind of certain they do “fair tests” where they test the same scenarios in different simulators. The problem however is that these tests are not indicative of anything useful. Why should one care if you get 430,000x real time speed simulating a robot arm not colliding or moving any objects with low sim accuracy? This number is fairly meaningless since if you do machine learning (specifically RL) to solve some task (manipulation, locomotion etc) that speed will drop significantly. Moreover in the past most simulators typically benchmark…

Paper and discussion history. A mention is not an endorsement.

Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning · HackerPapers discussions

Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning — mentioned by mmmore

Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning — mentioned by stonet2000